The present invention relates to methods and compositions for preparing antiviral nucleoside analogs, particularly BCH-189 (2',3'-dideoxy-3'-thia-cytidine). More particularly, the invention relates to the selective synthesis of the .beta.-isomer of BCH-189 and related compounds as well as the selective synthesis of enantiomerically-enriched BCH-189 and related compounds.
In 1981, documentation began on the disease that became known as Acquired Immune Deficiency Syndrome (AIDS), as well as its forerunner AIDS Related Complex (ARC). In 1983, the cause of the disease AIDS was established as a virus named the Human Immunodeficiency Virus type 1 (HIV-1). Usually, a person infected with the virus will eventually develop AIDS; in all known cases of AIDS the final outcome has always been death.
The disease AIDS is the end result of an HIV-1 virus following its own complex life cycle. The virion life cycle begins with the virion attaching itself to the host human T-4 lymphocyte immune cell through the bonding of a glycoprotein on the surface of the virion's protective coat with the CD4 glycoprotein on the lymphocyte cell. Once attached, the virion sheds its glycoprotein coat, penetrates into the membrane of the host cell, and uncoats its RNA. The virion enzyme, reverse transcriptase, directs the process of transcribing the RNA into single stranded DNA. The viral RNA is degraded and a second DNA strand is created. The now double-stranded DNA is integrated into the human cell's genes and those genes are used for cell reproduction.
At this point, the human cell carries out its reproductive process by using its own RNA polymerase to transcribe the integrated DNA into viral RNA. The viral RNA is translated into glycoproteins, structural proteins, and viral enzymes, which assemble with the viral RNA intact. When the host cell finishes the reproductive step, a new virion cell, not a T-4 lymphocyte, buds forth. The number of HIV-1 virus cells thus grows while the number of T-4 lymphocytes decline.
The typical human immune system response, killing the invading virion, is taxed because a large portion of the virion's life cycle is spent in a latent state within the immune cell. In addition, viral reverse transcriptase, the enzyme used in making a new virion cell, is not very specific, and causes transcription mistakes that result in continually changed glycoproteins on the surface of the viral protective coat. This lack of specificity decreases the immune system's effectiveness because antibodies specifically produced against one glycoprotein may be useless against another, hence reducing the number of antibodies available to fight the virus. The virus continues to grow while the immune response system continues to weaken. Eventually, the HIV largely holds free reign over the body's immune system, allowing opportunistic infections to set in and ensuring that, without the administration of antiviral agents and/or immunomodulators, death will result.
There are three critical points in the virus's life cycle which have been identified as targets for antiviral drugs: (1) the initial attachment of the virion to the T-4 lymphocyte, or macrophage, site, (2) the transcription of viral RNA to viral DNA, and (3) the assemblage of the new virion cell during reproduction.
Inhibition of the virus at the second critical point, the viral RNA to viral DNA transcription process, has provided the bulk of the therapies used in treating AIDS. This transcription must occur for the virion to reproduce because the virion's genes are encoded in RNA; the host cell reads only DNA. By introducing drugs that block the reverse transcriptase from completing the formation of viral DNA, HIV-1 replication can be stopped.
Nucleoside analogs, such as 3'-azido-3'-deoxythymidine (AZT), 2',3'-dideoxycytidine (DDC), 2',3'-dideoxythymidinene (D4T), 2',3'-dideoxyinosine (DDI), and various fluoro-derivatives of these nucleosides are relatively effective in halting HIV replication at the reverse transcriptase stage. Another promising reverse transcriptase inhibitor is 2',3'-dideoxy-3'-thia-cytidine (BCH-189), which contains an oxathiolane ring substituting for the sugar moiety in the nucleoside.
AZT is a successful anti-HIV drug because it sabotages the formation of viral DNA inside the host T-4 lymphocyte cell. When AZT enters the cell, cellular kinases activate AZT by phosphorylation to AZT triphosphate. AZT triphosphate then competes with natural thymidine nucleosides for the receptor site of HIV reverse transcriptase enzyme. The natural nucleoside possesses two reactive ends, the first for attachment to the previous nucleoside and the second for linking to the next nucleoside. The AZT molecule has only the first reactive end; once inside the HIV enzyme site, the AZT azide group terminates viral DNA formation because the azide cannot make the 3',5'-phosphodiester with the ribose moiety of the following nucleoside.
AZT's clinical benefits include increased longevity, reduced frequency and severity of opportunistic infections, and increased peripheral CD4 lymphocyte count. Immunosorbent assays for viral p24, an antigen used to track HIV-1 activity, show a significant decrease with use of AZT. However, AZT's benefits must be weighed against the severe adverse reactions of bone marrow suppression, nausea, myalgia, insomnia, severe headaches, anemia, peripheral neuropathy, and seizures. Furthermore, these adverse side effects occur immediately after treatment begins whereas a minimum of six weeks of therapy is necessary to realize AZT's benefits.
Both DDC and D4T are potent inhibitors of HIV replication with activities comparable (D4T) or superior (DDC) to AZT. However, both DDC and D4T are converted to their 5' triphosphates less efficiently than their natural analogs and are resistent to deaminases and phosphorylases. Clinically, both compounds are toxic. Currently, DDI is used in conjunction with AZT to treat AIDS. However, DDI's side effects include sporadic pancreatis and peripheral neuropathy. Initial tests on 3'-fluoro-2'-3'-dideoxythymidine show that its anti-viral activity is comparable to that of AZT.
Recent tests on BCH-189 have shown that it possesses anti-HIV activity similar to AZT and DDC, but without the cell toxicity which causes the debilitating side effects of AZT and DDC. A sufficient quantity of BCH-189 is needed to allow clinical testing and treatment using the drug.
The commonly-used chemical approaches for synthesizing nucleosides or nucleoside analogs can be classified into two broad categories: (1) those which modify intact nucleosides by altering the carbohydrate, the base, or both and (2) those which modify carbohydrates and incorporate the base, or its synthetic precursor, at a suitable stage in the synthesis. Because BCH-189 substitutes a sulfur atom for a carbon atom in the carbohydrate ring, the second approach is more feasible. The most important factor in this latter strategy involves delivering the base from the .beta.-face of the carbohydrate ring in the glycosylation reaction because only the .beta.-isomers exhibit useful biological activity.
It is well known in the art that the stereoselective introduction of bases to the anomeric centers of carbohydrates can be controlled by capitalizing on the neighboring group participation of a 2-substituent on the carbohydrate ring (Chem. Ber. 114:1234 (1981)). However, BCH-189 and its analogs do not possess a 2-substitutent and, therefore, cannot utilize this procedure unless additional steps to introduce a functional group that is both directing and disposable are incorporated into the synthesis. These added steps would lower the overall efficiency of the synthesis.
It is also well known in the art that "considerable amounts of the undesired .alpha.-nucleosides are always formed during the synthesis of 2'-deoxyribosides" (Chem. Ber. 114:1234, 1244 (1981)). Furthermore, this reference teaches that the use of simple Friedel-Crafts catalysts like SnCl.sub.4 in nucleoside syntheses produces undesirable emulsions upon the workup of the reaction mixture, generates complex mixtures of the .alpha. and .beta.-isomers, and leads to stable o-complexes between the SnCl.sub.4 and the more basic silyated heterocycles such as silyated cytosine. These complexes lead to longer reaction times, lower yields, and production of the undesired unnatural N-3-nucleosides. Thus, the prior art teaches the use of trimethysilyl triflate or trimethylsilyl perchlorate as a catalyst during the coupling of pyrimidine bases with a carbohydrate ring to achieve high yields of the biologically active .beta.-isomers. However, the use of these catalysts to synthesize BCH-189 or BCH-189 analogs does not produce the .beta.-isomer preferentially; these reactions result in approximately a 50:50 ratio of the isomers.
Thus, there exists a need for an efficient synthetic route to BCH-189 and its analogs. There also exists a need for a stereoselective synthetic route to the biologically active isomer of these compounds, .beta.-BCH-189 and related .beta.-analogs. Furthermore, there exists a need for a stereoselective synthetic route to enantiomerically-enriched .beta.-BCH-189 because the other enantiomer is inactive and, therefore, represents a 50% impurity.